Has anybody actually got any results from these processors please?

[anachronism]

Hi folks

Have any of the folks here actually managed to process these and get any results? There's all kinds of theories about them and I'd be very interested in hearing from anyone who has managed to get any progress.

 

[BShan]

I have no clue, but after doing some digging for about an hour for who-knows-why, I did find that these are likely IBM POWER5 MCMs (multi-chip modules) or similar. It looks like they're flip-chip, from other pictures of similar or older/newer POWER mcms and chips.

It has that purple ceramic which I've seen people theorize is a common indicator of beryllia. But also, I've only had one piece of beryllia, and it was white, so I take that claim with a hefty chunk of salt.

https://en.wikipedia.org/wiki/POWER5

Don't think we've talked, but I'm happy to see you here!

 

[anachronism]

Thanks BShan - there are also theories of vertical bonding wires- I'll look deeper into it. You're correct though on the type, they are all from IBM mainframes.

 

[kurtak]

Hi folks

Have any of the folks here actually managed to process these and get any results? There's all kinds of theories about them and I'd be very interested in hearing from anyone who has managed to get any progress.

Jon - there is not a lot of gold associated with the ceramic processors as there are no bond wires - just gold dots/pads on the under surface of the ceramic processor - VERY LOW yield

However - where the processor makes contact with the board there should be a plastic cage that has "fuzz buttons" in it which are relatively GOOD yield

Here is a trip down memory lane ----------

https://goldrefiningforum.com/threads/sun-ultra-sparc-cpus.16018/

What you show here is "a bit" different (multi-chip instead of single chip) so I am just "guessing " that because it is Sun Ultra Sparc ceramic technology the "fuzz button" connection is the same ?

Kurt

 

[CrystAG]

I pinched this quote from a Register article on the chip (for what it may be worth)

"The z196 processor has 1.4 billion transistors and weighs in with 512.3 square millimeters in real estate, making it a bit larger than the Power7 chip in both transistor count and area. The z196 chip uses IBM's land grid array packaging, which have golden bumps called C4 instead of pins. The z196 processor has a stunning 8,093 power bumps and 1,134 signal bumps."

https://www.theregister.com/2010/07/23/ibm_z196_mainframe_processor/

 

[anachronism]

Jon - there is not a lot of gold associated with the ceramic processors as there are no bond wires - just gold dots/pads on the under surface of the ceramic processor - VERY LOW yield

However - where the processor makes contact with the board there should be a plastic cage that has "fuzz buttons" in it which are relatively GOOD yield

Here is a trip down memory lane ----------

https://goldrefiningforum.com/threads/sun-ultra-sparc-cpus.16018/

What you show here is "a bit" different (multi-chip instead of single chip) so I am just "guessing " that because it is Sun Ultra Sparc ceramic technology the "fuzz button" connection is the same ?

Kurt

~Oh yes Kurt the fuzzies rock. I've got all those!

 

[g_axelsson]

Hi Jon, nice to see you here again!

Isn't that the same processor modules you had a box full of when I visited you? I got a sample with me and the result was only thin gold plating on the surface. The wiring inside consisted of copper wires, including the vertical vias.

Bond wires are connected through a bonding process where each end is fused to a connection point and is done in free space. In ceramic IC:s you always find a cavity where the bonding is made and protected by a lid. In EPROM you can even see the bond wires through the glass and in plastic capsules the bond wires are protected by the plastic when the body is moulded after the chip is bonded to the lead frame.

If the ceramic is undamaged it can be sold to collectors for way more than the gold content. These huge chips are popular with the collectors.

Göran